When this site gets something wrong, the fix is logged here, in public, with a date. A site that hides its corrections is not more reliable, only quieter about its mistakes.
14 September 2026 (continued)
The STS-80 bench said that at a full one-second thruster pulse the orbiter's delta-v "stays at parts of a millimetre". By the bench's own thrust and mass it is about 35 mm/s, a few centimetres a second; the flake's 20 m/s and the default 0.08 s figures stand.
The U-2 and OXCART bench said the glint "runs from magnitude -13 to -4 and never once falls below Venus", a sentence that contradicted itself, since Venus is -4.6 on the bench. At a hundred kilometres the glint stays brighter than Venus until the roughness passes about 15° and drops below it only at the top of the dial, outside the plausible band; the page now says so.
The LGM-1 bench called Vela "the second supernova-remnant pulsar, the southern confirmation that the Crab was no fluke", and its tier chip called it the second witness. The order was the other way round: Large, Vaughan and Mills announced the Vela pulsar in its remnant in Nature on 26 October 1968, and the Crab pulsar followed within weeks as the northern case. The preset, the chip and the panel button now say first, and the 1968 paper is in the sources.
The Rendlesham bench described its visibility dial as "a hundredfold swing in delivered light" between magnitude -10.2 and -7.4. Those two magnitudes are twelve times apart, and -7.4 is the dial's 10 km point, not its haze end; the whole dial, from 40 km to 5 km, spans some three hundred and fifty times, and the 20 km default sits about a hundred and fifty times above the haze end. Every place that said hundredfold now gives the bench's figure.
The Tic Tac bench introduced its 120-second preset as "under 0.1 g of net acceleration, transonic at worst". The bench's own brake profile for that preset is 0.24 g with a peak of Mach 0.44 at the midpoint, subsonic throughout, which is what its panel had been printing.
Three bench source lists, on the Radiation Dose, Skinwalker Spectrum and MiG-21 Disintegration pages, cited Future Visions as a 2023 or 2024 book by Lacatski, Kelleher and Knapp. The site's own Sources page and culture note have it right: James T. Lacatski alone, RTMA, 2026. The three citations and the two catalogue chips now match.
The Stealth bench said that during the Belgian wave "the two flying B-2s were daylight test aircraft at Edwards". Only one B-2 had flown: AV-1 first flew on 17 July 1989, and AV-2 did not fly until 19 October 1990, after the wave. The claim row, the try-this, the sources note and the file card now say one, and the argument they carry, that no stealth bomber was over Belgium, is unchanged.
The Kaikoura bench said Maccabee's 217,000 candela at 18 km "lands within his own rounding" and that his 18-to-35 km pair "matches to two percent", and priced the 6 km image at about 12,000 candela. Run on his published inputs, the bench's own equation gives about 260,000 candela at 18 km, roughly a fifth above his figure, with the same offset on his 35 km value, so the offset sits in one of the published constants; the ratio between his two distances does reproduce. The 6 km image prices at about 15,000 candela, and the pin now prints the computed value. The ray try-this said the refraction budget collapses "from a third of a degree to a hundredth while the gap stays at eight"; on the bench's own dials it goes from about five hundredths of a degree to about one hundredth, and the gap line holds near seven degrees after dip and refraction.
The Bolide Re-entry bench said the gap between 8.4 and 11.2 km/s is "deliberately dead air, because physics keeps it empty" and that a luminous entry "must hold one of the two bands". Decays from elongated Earth orbits arrive inside that gap: by vis-viva a GTO upper stage meets the air near 10.2 km/s and a lunar-return capsule near 11. The bands are now described as a strong discriminator, not an exclusive one, and the readout names the gap instead of calling it unnatural. The integrator's gravity term also carried the wrong sign for the code's own angle convention; the published figures it printed did not move at their stated precision, and the tune now checks the step against an independent integration.
The King's Chamber Resonance bench said the hydrogen line arrives at Dunn's section "four ways" and that squeezing the duct's width below about 6.9 inches makes the line "travel alone". The TE01 cutoff depends on the height alone, so at 4.8 inches no width on the dial gives a single mode; the width only strips the width modes, and the flip needs the height under 4.15 inches, half the wavelength. The section also carries five modes at the line, since TM11 shares TE11's cutoff. The (2,1,0) room mode prints 46.3 Hz, not 46.4, and the story's "noisy-day ceiling is one watt" has been split into the bench's own 1.3 W busy-site and 16 W noisy-site figures.
The Tiamat bench attached Δ(1/a) of 0.63 AU⁻¹, "a thousand times the nudge", to the single braking impulse. That figure is the circle-to-circle change; the one impulse gives about 0.17 AU⁻¹, about 1,600 times the nudge, and the full circle 0.63, about 6,000 times, the rest being the second burn. The intro, the story chapter and the readout now pair each figure with the manoeuvre it belongs to.
The Gimbal bench's panel listed "ALT 25,010, 0.58 M, the azimuth counting 54 L down to 7 L" as the numbers printed on the frames. Those are GoFast's altitude and Mach; the Gimbal frames read ALT 25,000 and CAS 240, and the azimuth crosses zero to 7 R, which the same panel had said elsewhere. Found by the calibration scripts now reading every copy location of a bench, not only its strings file.
The Trans-Medium bench's dial tip said to "set speed above ~250 kt and watch the path turn teal" as the cavitation number drops below the supercavitation threshold. At the panel's shipped dials the integrator does not open the cavity until about 320 kt, and at 250 kt nothing turns teal. The tip now says 320.
The Forbidden Science 7 note credited the paper on the 1966 Haynesville, Louisiana light event to the materials collaboration with Garry Nolan and Larry Lemke and dated it 2024. The paper is Vallée, Dini and Mestchersky, Progress in Aerospace Sciences vol. 156, 2025; Nolan and Lemke are not on it. The note, its reference list and the Haynesville 1966 and Larry Lemke wiki entries now say so, and the same note's list of official bodies now runs AAWSAP then AATIP, the order the record gives.
The Closer Encounters review quoted the book giving the Trickster the octopus as its avatar, "borrowed from the future". The book's phrase is itself a quotation from a 2018 paper by Edward Steele and colleagues in Progress in Biophysics and Molecular Biology: octopus genes that "seem to be borrowed from a far distant 'future' in terms of terrestrial evolution". The review also said the book ends on three words; the closing sentence continues: "We are the aliens. We are the Wings of an Other World, the World of Tomorrow." Both passages now quote the book as printed, and the paper is in the references.